It is very difficult to take issue with the pessimistic tone of Adam White’s sensible advice to the next President on 10 ways to promote the rule of law. All of the topics that he mentions are understood as serious, systemic weaknesses. When it comes to administrative law, President Obama has a penchant for excessive regulation on substantive matters. As will become clear, I differ with White on many specific points, but we agree on one central point. One of the reasons why the rule of law has fared poorly under the Obama administration is that the President shows little willingness to accept its requirements as a powerful or useful constraint on his various powers. Instead he operates under a program that is both autocratic on the one hand and misguided on the other.

President Obama starts, of course, with the substantive positions on which there is only one right answer—his own—and then games the system in order to find the best way to achieve that result. If he thinks that he has Congress with him, he will work through the traditional legislative channels. If he thinks that he does not have the Congress at his beck and call, then he will gin up some argument to show that he is entitled to bypass a Congress that has failed to go along with his vision of the just society.

In some cases, I sympathize with the President’s aims, as I did with at least some portions of his immigration program, which he has sought to administer by fiat. In other cases, I think his proposals are deeply unwise, including his ill-fated nuclear deal with Iran, which seems only to embolden that nation to continue its attacks against us and our allies, funded in part by billions of dollars of assistance that we pay them in used dollar bills.

Ambiguities in the Scope of Executive Power

Yet it would be a deep intellectual mistake to think that any disagreement with a sitting President necessarily means that he has flouted the rule of law. It is quite clear that, for high-level executive actors, some discretion is a key portion of the job. Thus the ultimate challenge is to figure out when the President’s use of executive power should be condemned and when it should be tolerated. When stated in this form, the problem is acute. The textual command of immediate interest is from Article II, Section 3, which provides that the President “shall take care that the Laws be faithfully executed.”

The initial impulse from reading Section 3 is that the chief executive’s job is to carry out the will of Congress as it is expressed in legislation—a view that comports with the basic constitutional contrast between legislative and executive power.  That line depends on two key structural features. The first of these is the separation of powers, which entrusts particular tasks to particular branches of government, such that any branch that goes beyond its enumerated powers has flouted the basic constitutional commands. The second is the related notion of checks and balances, by which each branch is in a position to stop, in some settings at least, the unilateral action of a coordinate branch of government.

The combination of these two principles receives its most compelling application in the method by which laws are made. Congress enacts laws, but is subject to the check of a presidential veto, which in turn can be overridden by a two-thirds vote in each house of Congress, starting with the house in which the bill originated. The clear sense is that this process is justified by a presumption, under constant attack today, against the enactment of new laws. Why should infrequency be the default setting? Because of the risk to legislative politics posed by the problem of faction.

Yet in order for this model to work, everyone must have some confidence that the laws already in place are clear enough for us to know what they require and what they proscribe. As the scope of legislation becomes more complex, this assumption becomes more questionable for at least two reasons.

First, there are many statutes that consciously delegate to the President the right to initiate certain kinds of managerial decisions. In this regard, the position of a corporate CEO is instructive, because there are cases in which a company’s board of directors either expands or limits its CEO’s purview. Whenever Congress wants the President to decide certain questions—for example, where to locate U.S. post offices in particular towns or how to run the military—it can hardly be described as a breach of the duty to faithfully execute the laws to fill in the gaps that Congress believed were required by sensible administration. Sometimes the faithful execution of the law demands the exercise of intelligent discretion because no general legislation can foresee all the unanticipated contingencies that arise after implementing any legislative program.

At this point, courts and scholars alike commonly resort to the tripartite test that Justice Robert Jackson advanced in Youngstown Sheet & Tube Co. v. Sawyer, a 1952 Supreme Court case that rejected President Truman’s bid during the Korean War to unilaterally seize certain steel facilities when they were idled by a union dispute. The Court held that the President did not have this power, even though he was surely commander-in-chief of the Army and the Navy, whose effectiveness depended on a secure source of steel.

Noteworthy in those anxious deliberations was Justice Jackson’s concurring opinion that divided the terrain into three parts: the President’s power is at its greatest when the action was authorized by Congress; it is somewhere in the middle when Congress was silent; and it is at its “lowest ebb” when the actions were incompatible with the expressed or implied will of Congress. This formulation creates a useful guideline, but it hardly answers all cases. Some statutes are so unclear that it is hard to know what sentiment they express. Some statutes are so inscrutable that it is hard to draw implications from their silence.

The argument here is not that this tripartite division is useless. Often it is possible to find a sensible reading of the congressional tea leaves. But consider that a huge portion of the current judicial docket of the Supreme Court asks just how much deference should be afforded to the President or to administrative agencies when those agencies interpret the law. The cautionary note that follows from this evident difficulty is that Congress ought, in the first place, to pass laws that avoid the unexplored complications of statutory interpretation.

It is from this vantage point that we start to see both the substantive and the procedural dimensions of the rule of law. There is little doubt historically that the emergence of rule-of-law values was closely associated with efforts to rein in legislative and executive power. Procedurally, such limitations are achieved by making the statutes sufficiently clear to the President, or to the people whom he supervises—remember, the law requires that the President take care to see that the laws be faithfully executed, that is to say, often by someone else.

One way to avoid legislative ambiguity is to move toward modesty in legislative objectives, and the best way to do this is by creating systems of clear property rights that minimize the need for administrative discretion. An illustration of what happens when legislative modesty is disregarded starts with the Radio Act of 1927, which morphed into the more expansive Federal Communications Act of 1934.

The initial impetus for the regulation of the broadcast spectrum was the chaos that arose in the early 1920s when all broadcasters were allowed to broadcast on all frequencies at any power, so that no single broadcast could be heard above the din. Creating a system of property rights could stop that interference, and there were imperfect efforts to do that by assigning frequencies to the first person that occupied them. But that system soon foundered because the then-nascent federal law that governed the spectrum did not allow the government to limit where any eager entrant could broadcast or what power that broadcast signal could be.

The simplest way to have avoided that problem would have been for the United States to set up property rights in different segments, which could then be auctioned off to potential users for the highest price. The bundle of rights was defined by frequencies, where the key condition was minimizing interference at the margin between the users of neighboring frequencies. Once the U.S. government set the boundaries, the owner of that band was entitled to use it in whatever way he saw fit, and had the power to mortgage, sell, or subdivide, so long as these transactions respect the boundaries. At that point, updating the use of the spectrum fell—or should have fallen—to private parties, who could respond to market pressures and technological changes. In such a system, the state would only be needed for the (ordinarily) common law tasks of policing boundary lines and enforcing contracts.

But no; starting with the Radio Act, the system went off in different direction. The need to control physical interference was leveraged into developing a statutory scheme that let the government allocate broadcast frequencies in accordance with “the public interest, convenience and necessity,” which turned out to be a standard that invited discretion in assigning frequencies to different applicants. At that point, courts knew that they could not take the lead in exercising that discretion by themselves. But since they were committed to the administrative state, they let the Federal Communications Commission determine not only the rules of the road but also the composition of the traffic—at which point there were no obvious limits to the kinds of considerations that were relevant, even though none of them, alone or in combination, was ever decisive in making the allocation. Thus has the size of the administrative state mushroomed, but the allocative decisions are much inferior to those that would have been made if the frequencies had been auctioned off. The winning bid gives an accurate estimate of future value. The administrative hearing, in contrast, favors parties with glib presentations that are costly to present and impossible to evaluate.

This overconfident attitude toward administrative action carries over into modern times, one reason why the FCC is inclined to impose the regulatory mare’s nest of net neutrality, whose consequences are likely to be: raising costs, lowering revenue, and eventually less investment in telecommunications. The same result follows for any administrative agency that seeks to “improve on” common law rights of property and contract rather than finding better ways to clarify and enforce those rights. Thus the Environmental Protection Agency should concentrate its efforts on pollution control, not preserving habitats. The EPA should buy the habitat it needs either in a voluntary market or, if need be, by condemnation at a fair market value price. Anything that shrinks the scope of the administrative state to essential government functions will help to limit its mischief.

White’s Suggestions for Taming the Administrative State

At this point, my reaction to the White proposals should become clear. Pruning excessive regulation is surely the top priority. But in most cases it is not just a question of doing less by regulation. It is also, in a wide range of areas, a question of doing the right tasks by regulation. We must take care how we run the administrative state. On this score, it is a mixed blessing to seek ways to develop a unified agenda for the administrative state. Putting all the eggs into one basket is fine if the proposals that come out are fine. But it is a risky business if the unified agenda leads to more effective enforcement of dubious laws. The strong procedural commitment here carries with it certain real institutional risks.

In a similar vein, it is surely the case that running OIRA (the Office of Information and Regulatory Affairs) better is valuable. It is always necessary to coordinate inconsistent regulatory schemes. But there is a real danger that OIRA will take it as its mission finding effective ways to enforce dubious statutes rather than looking for ways to get the government out of certain lines of business—such as labor-market regulation, for which competitive forces do a far better job of matching human capital with employers’ needs.

It is also difficult to decide whether we want the President to exercise greater control over the independent administrative agencies or whether we want to take them out of his orbit. It is not clear which option leads to more sensible decisions. Thus, one reason for the net neutrality breakdown was that President Obama crossed a historically firm red line when he pressured Thomas Wheeler at the FCC to adopt the net-neutrality rules. The FCC has a five-member commission, deeply divided by politics, such that when the President uses his muscle, the usual and salutary internal deliberation breaks down.

Yet just recently a panel of the Circuit Court for the District of Columbia in PHH Corporation v. Consumer Financial Protection Bureau struck down a statute that empowered the head of that agency to unilaterally affect a wide array of financial decisions while being subject to only limited judicial review. The Circuit Court said that fundamental principles of separation of powers were violated when these powers were concentrated in the hands of a single person (the director of the CFPB). In its view, the Congress has a choice: It can either vest the power of the independent board in multiple commissioners, or it must subject a single commissioner to the absolute removal power of the President. Apparently either scheme is permissible, which gives some evidence as to the uneasiness that any court has about undoing, especially at this late date, the constitutional status of all independent agencies.

There is good reason for the hesitation. It is far from clear to me that the best solution is to let any President either dictate or substantially influence how this or other independent agencies with multiple commissioners can behave. I don’t think anyone would trust the President with the broad powers that are now delegated to these agencies. Yet it is just that outcome that informal influence, not easily subject to judicial oversight, creates. Hence, if agency independence is to be lessened, it should never happen by ad hoc presidential interference, but by political and even constitutional reconsideration, as a high-profile item of policy debate in Washington and in the country at large.

I have much the same concern about the use of administrative guidance, especially if it cannot be challenged in court prior to implementation. In some cases, the guidance offers welcome information on how to proceed. But in other cases guidance represents a crude form of coercion, allowing the government to extend its reach without going through notice-and-comment proceedings that might result in better rules. But the choices here are, alas, in the domain of the second-best, because all too often what is needed is a substantive repeal of rules that block individual choice.

Think of the serious danger of abuse on the part of the U.S. Food and Drug Administration that is invited by directives to keep much-needed, if high-risk, drugs off the market. The government seems to think that its experts know more than patients (and the parents of patients) and their physicians about which courses of treatment make sense and which do not. But the bulk of knowledge for individual cases surely lies with the patient and his or her physicians, who can combine generic information about prescription drugs with specific information about the patient’s condition.

It is only the constant application of FDA powers that dulls people to the simple insight that that agency is extraordinarily intrusive when, by reducing people’s options for treatment, it keeps them from taking prudent risks. White’s suggestions on refining administrative procedures cannot have the desired effect. In this instance, it is far better to try to reduce the administrative purview of the FDA than to encourage better use of those procedures, in, for example, dealing with clinical trials and drug approvals. White’s approach is too sympathetic to the guidance given the recent history of its widespread abuse. Obviously, all guidances cannot be eliminated in any complex system. But there is no reason to deny any party that is subject to harsh rules the right to challenge them immediately in court, without having to run the risk of serious punishment by acting in ways that could attract government enforcement.

Needless to say, I think there is much to be said for White’s proposal to reduce the scope of what I have termed “government by waiver.” Yet it is important to understand the legal forces that make this practice so difficult to rein in. Current law allows the government to impose all sorts of legal restraints on the ability of employers, for example, to contract with their employees over healthcare plans. In some instances, the regulatory burden of healthcare laws is so great that the firm will drop all coverage unless it can reduce the burden. So the administrator who fears a mass exodus from some important market allows for the waiver that in the short run at least is needed to keep the market afloat.

But that initial decision to use waivers soon raises questions such as: Who gets the waiver? What terms are waived?  How long does the waiver last? Can it be renewed? The questions can have a political valence. Are the waivers more common in Blue than in Red states? Are they more common for unions than for employers? Is there consistency across states and regions? These waivers all affect competitive balance, and yet the competitors who don’t get the waiver have no standing in court to challenge it. So the system just festers.

It is not sufficient to say don’t be “capricious” in granting waivers, or be more “transparent” in how they are granted. Of course consistency is a virtue of sorts, but in the regulatory quagmire, there are so few like cases that it is hard to know how much discipline is brought to bear by requiring consistency. Hence we should think hard about the first-best problem. Why create statutes that need waivers to survive? The first-best solution can only be reached in a regime that accords far greater respect to freedom of contract than the current law.

White is emphatically correct, moreover, when he attacks the Obama administration for diverting funds to its preferred political ends in the absence of congressional authorization. The Justice Department’s bank settlements are clearly one illustration of that about which I have written previously. The technique in question is this: The DOJ offers the banks a deal in which they will get credit against their fines if they make donations to certain activist groups. The credits given are more advantageous for the banks than is paying down their federal fines directly, so they have every incentive to accept these deals. Indeed, it would be hard, given their fiduciary duties to their shareholders, not to accept them. But the simple point is that the DOJ should never make these deals at all.

The same argument backs White’s criticism of the administration’s decision to divert receipts that should go to the U.S. Treasury to instead compensate insurance companies for their incurred losses under Obamacare. The case illustrates the infectious nature of expansive government programs. The excessive demands created by one program force the need to ignore legal constraints tied in with a second program, and so on down the line, until a culture of financial corruption pervades public administration. It is very hard to stop these problems at the second step. It is best therefore always to remember that the passage of any major legislation has negative collateral consequences outside the direct sphere of its application.

The next of White’s prescriptions, number eight, also deserves our strongest endorsement because it again goes to probity in public administration. There is an obvious sense of public malaise, for example, at the special treatment that Hillary Clinton and her various assistants have received in the FBI probe of her private email server.

One of the worst features of this sorry episode was the decision by Attorney General Loretta Lynch to defer to the decision of FBI director James Comey on whether or not to prosecute Secretary Clinton for various offenses relating to her emails. That decision was wholly inconsistent with the well-established roles of the DOJ and the FBI in criminal investigations. Once Lynch found herself compromised by meeting with Bill Clinton, she had to recuse herself from the case. But at this point she should have turned the matter over to someone else in the DOJ to assume her duties. She also should have guaranteed that person’s independence and final authority over the matter. As White noted, this is not the same as the creation of special prosecutors outside the DOJ, which are subject to the abuses that the late Justice Scalia noted in his prescient dissent in the 1988 case of Morrison v. Olson. Yet at the same time, it is not clear that the President should have any say in these actions, either, given his close political attachment to his former secretary of state. Conflicts of interest are fiendishly difficult to solve even in the most basic cases. They don’t get any easier when all the main actors appear to be driven by strong partisan motivations given the high stakes in the political system.

The same issues arise of course with the moral leadership of the President in connection with law enforcement. The great tragedy of Ferguson, Missouri was the administration’s utter silence when the legal case against Officer Darren Wilson was concluded, even when the DOJ report cleared the officer of any wrongdoing. At that point, President Obama, along with then-Attorney General Eric Holder, should have made it blindingly clear to everyone that, contrary to their own preconceptions, the police had done nothing wrong. Their disingenuous silence only fueled ever greater disrespect for the police, which in fact made their own task more difficult in the ensuing months, and may well have  contributed to the recent rise in homicide rates, especially in some major cities like Chicago.

The President has to admit his own errors and protect public officials who have done their jobs. The theory of small government only requires that we limit the tasks given to public officials. It does not call for weak and indecisive leadership on matters of public safety, which is what the Obama administration has provided. The result of that weakness has been a massive deterioration in race relations, which the President could have, and should have, done much to avert.

And yes, do get the oath of office right, as White suggests. That is the first step in a long journey to regain a correct understanding of the key role of presidential power in preserving the rule of law, and the interests in individual liberty, private property, and contractual freedom that the presidency should strive to protect.